Choose where AI runs
- System AI
- Use the AI provided by your operating system when it is available. Availability depends on the device and OS.
- Local LLM
- Use the existing QuickLLM, Ollama or LM Studio settings with a server you run on this device or your network.
- Custom Server
- Connect using a Base URL, model ID and, when required, an API key. This can be your own server, a hosted API or a gateway.
See OpenRouter, OpenAI, Anthropic and Gemini configuration examples
Open AI settings
- Open Settings in the QuickApps app you want to use, then open its AI settings.
- Select an existing local provider, or add a Custom Server.
- Enter the connection details and use Save & Verify. This sends a short chat request and a hosted API may charge for it.
- Select the connection in your chat. For summaries, suggestions or other one-shot actions, also check AI action provider.
Adding a server does not automatically change every chat or the AI action provider. The shared settings interface does not mean API keys or configuration are automatically shared between apps.
Keep local connections available
Existing local providers remain available. For a server on another computer, use its reachable LAN address. On a phone, localhost refers to the phone itself. In QuickLLM, use the server address, client key and Profile alias shown for that connection.
Privacy and availability
Text or other content sent for an AI action goes to the selected server. A gateway may forward it to its model provider. Check the provider’s terms, data handling and supported countries before sending sensitive information.
This website does not ask for your API key and does not run an AI request. Guide availability is independent of using the app’s settings.